Build an Index on Scalars
This guide describes how to build an index on scalar fields.
Overview
Unlike vectors, which have both magnitude and direction, scalars have only magnitude. Milvus regards single numbers and strings as scalars. Here is a list of the available data types for scalar fields in Milvus.
To speed up attribute filtering in hybrid searches, you can build indexes on scalar fields since Milvus v2.1.0. You can read more about scalar field indexing here.
When interacting with Milvus using Python code, you have the flexibility to choose between PyMilvus and MilvusClient (new). For more information, refer to Python SDK.
Build index
To build an index on scalar fields, you do not need to set any index parameters. The default value of a scalar field index name is _default_idx_<fieldId> followed by the name of the indexed field. You can set it to another value that seems fit.
The following code snippet assumes that a collection named book already exists and an index is to be created on the string field book_name.
from pymilvus import Collection
collection = Collection("book")
collection.create_index(
field_name="book_name",
index_name="scalar_index",
)
collection.load()
Once the index has been created, you can include a boolean expression on this string field in a vector similarity search as follows:
search_param = {
"data": [[0.1, 0.2]],
"anns_field": "book_intro",
"param": {"metric_type": "L2", "params": {"nprobe": 10}},
"limit": 2,
"expr": "book_name like \"Hello%\"",
}
res = collection.search(**search_param)
What's next
To learn more about scalar field indexing, read Scalar Index.
To learn more about the related terms and rules mentioned above, read